Dogs as Emotion Readers: What Your Dog Shows You About Yourself

In brief

Dogs respond finely to measurable signals from their people – facial expressions, voice, posture, even scent – and adjust their behavior accordingly. In that sense, your dog can reflect a great deal about your own state back to you: your tension, your consistency, your patience.

Important context: “Mirror” is a helpful image, not a measuring technique. Some of the effects described here are well documented – for example, that dogs can tell facial expressions apart. Others are correlational, such as the similarity in cortisol levels between people and their dogs, and do not allow simple cause-and-effect conclusions. And the fact that a dog distinguishes emotional signals does not automatically mean that he feels them the way we do. The thoughts on self-reflection in this article are meant as food for thought, not as a diagnosis.

Dogs are more than companions. From the perspective of behavioral biology, neurobiology and developmental psychology, they are exceptional social partners whose adaptation to humans is remarkable in its own right. They are not merely receivers of human signals but active participants in a reciprocal feedback system.

1. What this article is about

1.1 The actual question

Not whether dogs “understand” us. Rather: Which signals do they demonstrably perceive, what follows from that for their behavior, and what can you use for your own reflection without reading more into it than is there?

1.2 How to read the evidence

The key studies in this field work with a few dozen animals, often from particular breeds or groups of owners. Several of the findings mentioned here are also disputed in the scientific debate or were corrected after publication. This is noted at the relevant points.

1.3 What this article does not claim

It does not claim that the dog is a measuring instrument for his person’s mind. And it does not claim that perceiving is the same as empathizing.

2. Cross-species emotion recognition: the sensory basis

2.1 Facial expressions

That dogs can distinguish human emotional signals has been well studied experimentally. Müller and colleagues (2015) at the Messerli Research Institute in Vienna trained dogs on a touchscreen to discriminate happy from angry human faces – using only the upper or only the lower half of the face. The dogs then transferred what they had learned to unfamiliar faces and to the other half of the face. This made it possible to rule out that they were simply following basic features such as visible teeth (10.1016/j.cub.2014.12.055).

Two things matter here. First, this was not untrained recognition but a clean demonstration that the emotion itself can serve as the feature for discrimination. Second, the group trained to choose happy faces learned considerably faster – an indication that the dogs already brought prior experience with them.

2.2 Multimodal integration

Albuquerque and colleagues (2016) showed that dogs do not just process emotional signals separately through eye and ear but bring them together: facial expression and tone of voice are integrated into a coherent assessment (10.1098/rsbl.2015.0883). This ability had previously been considered established mainly in primates.

2.3 What can be measured in the brain

In an fMRI study with eleven dogs lying awake in the scanner, Andics and colleagues (2014) showed that the dog brain – like the human brain – has voice-sensitive regions that respond to voices and their emotional content (10.1016/j.cub.2014.01.058). An important distinction: this study concerns the processing of voices, not faces. Whether, and to what extent, dogs have dedicated face processing is the subject of further research, some of it still inconsistent.

2.4 Through scent

Dogs also perceive chemical signals from humans. D'Aniello and colleagues (2018) presented Labrador and Golden Retrievers with armpit sweat samples collected from male donors under fear and happiness conditions. With the fear odor, the dogs showed more stress behavior and a higher heart rate; with the happiness odor, they turned more toward a stranger (10.1007/s10071-017-1139-x).

Wilson and colleagues (2022) took a different approach: they specifically trained dogs to discriminate between odor samples taken from the same people once at rest and once under acute psychological stress. Under controlled conditions, the dogs succeeded at this discrimination (10.1371/journal.pone.0274143). Note the difference: what was demonstrated here is that the odors can be told apart, not spontaneous recognition in everyday life.

The idea that dogs routinely “read” our cortisol levels or our heart rate variability goes beyond these findings. What is documented is perception through scent, facial expression and body language. How scent processing drives behavior is covered in The World Through the Nose: How Olfactory Processing Drives Canine Behavior.

2.5 The evolutionary picture

Dogs’ social attentiveness is considered the result of a long process of domestication. Hare and colleagues (2002) compared dogs, human-raised wolves and great apes in a task in which the location of hidden food could only be inferred from a human gesture. Dogs used the gesture more reliably than wolves and great apes, and even puppies only a few weeks old with little human contact could do it (10.1126/science.1072702). How far this goes: the interpretation that this is the result of selection specifically for social cognition is the research team’s interpretation – alternative explanations such as early learning experience are still debated today. More on this in The Domestication Syndrome: What the Evidence Supports in Dogs.

2.6 For self-reflection

Dogs do not respond to some mysterious “aura” but to concrete signals – many of which we send without being aware of them. If you learn to read your dog’s behavior as feedback, you get clues about your own tension, consistency and body language. A clue, not a verdict.

Further reading: How dogs perceive and respond to our emotions

3. The dyad as a feedback system

3.1 Behavior does not run in one direction

Behavioral biology speaks of a bidirectional feedback process: behavior produces responses, which in turn influence the original behavior. You can observe this as mutual adjustment – dogs align their movement and attention with their people, and people unconsciously adapt their posture and voice to the dog. This is a well-described everyday observation; how strong this adjustment is in an individual case depends on the relationship, the situation and training.

3.2 Oxytocin: the best-known and the most disputed finding

Let’s put it carefully: certain forms of human-dog interaction can be accompanied by changes in the oxytocin system. Nagasawa and colleagues (2015) reported that longer mutual gazing between dog and owner was associated with higher urinary oxytocin concentrations in both, and that after intranasally administered oxytocin, more gazing was observed in female dogs, which was also reflected in the owners. This pattern did not appear in human-raised wolves (10.1126/science.1261022).

The common shorthand “eye contact releases oxytocin” goes beyond this. What was measured were urinary concentrations and their relationship to observed behavior – not a process triggered in the brain, and certainly not a feeling.

This study is often cited as evidence of a “love hormone loop.” In the scientific debate it is disputed: Fiset and Plourde (2015) raised methodological objections in a published commentary, among other things regarding the participation of researchers with their own dogs and the question of whether the conclusions about coevolution are supported by the design (10.3389/fpsyg.2015.01845). The finding stands; the far-reaching interpretation has not gone unchallenged. In detail in Oxytocin in Dogs: Bonding, Trust, and Stress Regulation.

3.3 For self-reflection

The quality of your life together can make something visible about your own self-regulation and the way you shape relationships. This is an invitation to reflect, not an objective measure of your social competence.

How to strengthen this relationship deliberately: Building a stronger bond with your dog

4. Cortisol in step

4.1 What was measured

At Linköping University, Sundman and colleagues (2019) studied 58 human-dog pairs – 25 Border Collies and 33 Shetland Sheepdogs, all with female owners. Hair cortisol was measured at two time points. This comes with a limitation that is usually missing in popular accounts: hair cortisol reflects longer-term patterns of strain over weeks and months, not the short-term stress response in a single situation. Anyone who reads this as the dog reacting to his person’s mood on a particular day is confusing two different timescales. The result: a significant positive correlation between the values of dog and owner (10.1038/s41598-019-43851-x). An author correction to this paper was published later (10.1038/s41598-020-74204-8).

4.2 How to read the finding

• The dogs’ physical activity did not explain the values. But that does not mean that “emotional closeness” causes the synchronization – it only rules out one explanation.
• The dogs’ personality was barely related to their cortisol levels, whereas the owners’ personality clearly was: neuroticism, conscientiousness and openness were significantly associated with the dogs’ cortisol. From this, the authors cautiously conclude that it is more likely the human who affects the dog than the other way around – and they explicitly stress that nothing certain can yet be said about the cause of the correlation.
• So it is an association, not a proven transfer. The sample – two herding breeds, female owners only – considerably limits how far the results can be generalized. Both breeds have been selected for close cooperation with humans; whether the same pattern occurs in breeds that work more independently was an open question at the time of the study.

4.3 How stress shows in dogs

Level	Possible signs
Behavior	Restlessness, displacement behaviors (scratching, licking), jumpiness, noise sensitivity
Autonomic nervous system	Panting, increased heart rate, trembling, drooling
Gastrointestinal	Diarrhea, vomiting, loss of appetite
Cognitive	Reduced attention, reduced ability to learn, increased reactivity

How to interpret cortisol in dogs and where measurement reaches its limits is explained in Chronic Stress in Dogs: Neurobiology, Cortisol and Long-Term Behavioral Impact. One important point: these signs are nonspecific. They can point to stress, but also to pain or illness – see Recognizing pain in dogs.

4.4 For self-reflection

The correlation suggests at least adding a second question to “What’s wrong with the dog?”: “What mood am I bringing with me?” The relationship probably works in both directions: if you take care of your own stress relief, you may be doing your dog some good as well. Seen this way, deliberate downtime is care for yourself and for your dog at the same time.

What emotional support can look like in practice: Comforting a fearful dog – is it okay?

5. Attachment theory in the human-dog dyad

5.1 The Strange Situation Test in dogs

Attachment theory, familiar from developmental psychology, has been applied to the human-dog relationship. Topál and colleagues (1998) were the first to apply the Strange Situation Test to dogs and showed that dogs display attachment behavior toward their people and use them as a secure base for exploring (10.1037/0735-7036.112.3.219).

Solomon and colleagues (2019) developed a classification system from this that stays close to Ainsworth’s procedure. For this, 59 adult dogs and their caregivers went through a mildly threatening laboratory encounter with a stranger as well as the Strange Situation procedure (10.1080/14616734.2018.1517812).

5.2 The patterns described

• Securely attached dogs: use their person as a secure base for exploring and appear more balanced.
• Insecure-avoidant dogs: orient little toward their person and avoid closeness.
• Insecure-ambivalent dogs: cling, are hard to calm down, and react strongly to separation.

In the Solomon study, 61 percent were classified as secure and 39 percent as insecure – similar to human toddlers. The caregiver’s sensitivity during the encounter with the stranger significantly distinguished secure from insecure dogs; in addition, a lower score on a personality scale for activity and excitability was associated with a secure classification.

The authors explicitly describe their work as exploratory. With 59 dogs and a classification applied for the first time, these are well-founded patterns, not established categories. In detail in Attachment in Dogs: How Early Experiences Shape the Human–Dog Bond.

5.3 For self-reflection

How we shape attachment – clear, supportive, insecure or intrusive – can help shape the dog and can make our own relationship patterns visible. As an invitation to reflect, not as a psychological diagnosis.

6. Impulse control as a shared task

6.1 What lies behind it

Impulse control is linked to the functions of prefrontal networks – in humans and in dogs. Reward-based training can work on several levels here: it practices behavioral inhibition through repeated choices, can promote self-regulation through delayed reward, and can reduce stress-related impulsivity through predictability. The neural attribution is plausible; the finer mechanics have not been mapped in every detail in dogs. More in Self-Control in Dogs: The Role of the Frontal Cortex, Stress, and Training and in Arousal Regulation in Dogs: Self-Control, Learning and Calm.

6.2 Training is a dyadic process

Whether it’s sit and stay, loose-leash walking or waiting before the food bowl – all of this demands impulse control not only from the dog but also from the human: precise timing, consistency and emotional regulation. Impatience can disrupt the learning process.

6.3 For self-reflection

Working with your dog in a structured way trains your own patience and consistency along the way. The dog gives immediate feedback: if you are inconsistent, you will often get inconsistent results; if you stay calm and clear, you can encourage calm and clarity.

Training approaches grounded in learning theory: Learning theory in dog training

7. The dog as an anchor for mindfulness

7.1 An observation, not a measurement

Dogs orient strongly toward current environmental stimuli and immediate experiences – that is a widespread, everyday observation. Treating it as an established statement about how dogs perceive time would go too far, though: there is hardly any solid research on this in dogs, and what is occasionally cited as evidence comes from studies of other species. The phrase “dogs live in the moment” is an attribution, not a measured quantity.

A good deal speaks against the absolute version anyway: the fact that dogs remember differently from humans does not mean they have no complex memories. See Episodic-Like Memory in Dogs: What Dogs Retain and for How Long.

7.2 For self-reflection

The dog asks for presence and rewards genuine connection. In an age of constant availability, that is a welcome counterbalance – it cannot be sped up or delegated. Part of the picture is that the effect on people varies greatly from person to person and is not automatic.

8. Projection: the dog as a reason to look closer

8.1 Why this section is different

It is deliberately interpretive, not empirical. How we interpret our dog’s behavior often reveals something about our own assumptions.

8.2 A concrete example

If you constantly see “dominance” in your dog, you are working with a hierarchical picture that modern behavioral research does not support in that form. The dog himself is not dominating anything here – the interpretation says more about the human’s mental model than about the animal. Why the concept has failed scientifically is explained in Why Dominance Fails as a Scientific Concept in Dogs. How a state differs from a behavior is covered in Behavior Does Not Equal Emotion.

8.3 What this means for us

The dog can confront us with our own patterns: where we are patient and where we are not, where we communicate clearly and where we send mixed messages. That is a valuable aid to reflection – as long as you take it as such and don’t inflate it.

9. Where the evidence comes from

The samples in this field are small: eleven dogs in Andics, 58 pairs in Sundman, 59 dogs in Solomon. D'Aniello used two breeds; Sundman used two herding breeds and female owners only. These are valuable clues, not general laws.

The fields are also concentrated: a considerable share of imaging and cognitive research on dogs comes from a small number of research groups in Budapest and Vienna. Dogs lying awake in a scanner are demanding to work with, and only a few labs manage it. Independent replications are largely lacking.

It is also striking how often the widely cited papers have not gone unchallenged in the scientific debate. There is a published methodological objection to the oxytocin study and an author correction to the cortisol study. This does not invalidate the findings – it shows that this is an ongoing research process, not a finished picture.

In fairness, two gaps should be named. First, the question of cause: the cortisol similarity is correlated, not causally explained. Second, the question of experience: all the methods mentioned here measure behavior, hormones or brain activity; none of them measures what something feels like for the dog.

10. What the research does (not yet) show

• Distinguishing is not feeling. That dogs distinguish facial expressions or voices demonstrates perception – not that they experience emotions the way we do.
• Correlation is not causation. The cortisol similarity is a correlation; the direction from human to dog is plausible but not proven.
• Long-term patterns are not day-to-day form. Hair cortisol captures strain over weeks and months. This finding says nothing about whether your dog picks up on your tension today.
• Voice is not face. The fMRI findings concern voice-sensitive regions; dedicated face processing in dogs is less clearly established.
• Scent and facial expression, not “reading cortisol.” Dogs perceive stress through scent, body language and facial expression – not by reading lab values.
• Trained discrimination is not spontaneous recognition. In the scent study, the dogs were trained for the task; that says little about their behavior in everyday life.
• Small, specific samples. Several key studies are based on a few dozen animals and particular breeds or groups of owners.
• “Mirror” is a metaphor. The dog responds finely to signals – he is not an objective measuring device for his person’s mind.

11. Conclusion: the dog as a prompt for your own development

From the perspective of behavioral biology, neurobiology and attachment theory, the dog is not a passive companion but an active part of a social system. What is documented: dogs distinguish human emotional signals and bring together what they see and hear; there is a measurable correlational similarity in stress between dog and human; attachment patterns can be described with adapted procedures; and training demands executive functions on both sides.

If you look closely, you can recognize a lot about your own state in your dog – about stress, impatience, relationship patterns. That makes this relationship not only emotionally enriching but also personally instructive. Just don’t inflate the mirror: the dog responds to real signals, but he is not a diagnostic instrument. If you are willing to reflect on your own behavior, you can grow in this dyad – in self-regulation, empathy and the way you shape relationships.

Key takeaways

• With training, dogs can discriminate happy from angry human faces and transfer this to unfamiliar faces and to the other half of the face (Müller et al., 2015).
• Facial expression and tone of voice are combined into a joint assessment (Albuquerque et al., 2016).
• The dog brain has voice-sensitive regions that respond to the emotional content of voices – the study concerns voices, not faces (Andics et al., 2014).
• Human fear odor increased stress behavior and heart rate in dogs (D'Aniello et al., 2018); trained dogs were able to discriminate between the resting odor and the stress odor of the same person (Wilson et al., 2022).
• In 58 pairs from two herding breeds, the hair cortisol levels of dog and owner were correlated; the owner’s personality was clearly associated with the dog’s cortisol, the dog’s own personality barely. Hair cortisol reflects longer-term patterns of strain, not a single stressful situation (Sundman et al., 2019).
• In an exploratory study with 59 dogs, 61 percent were securely attached; the caregiver’s sensitivity distinguished secure from insecure dogs (Solomon et al., 2019).
• Longer mutual gazing was associated with higher urinary oxytocin levels; the “gaze loop” derived from this has not gone methodologically unchallenged (Nagasawa et al., 2015; Fiset & Plourde, 2015).

References

Albuquerque, N., Guo, K., Wilkinson, A., Savalli, C., Otta, E., & Mills, D. (2016). Dogs recognize dog and human emotions. Biology Letters, 12(1), 20150883. https://doi.org/10.1098/rsbl.2015.0883

Andics, A., Gácsi, M., Faragó, T., Kis, A., & Miklósi, Á. (2014). Voice-sensitive regions in the dog and human brain are revealed by comparative fMRI. Current Biology, 24(5), 574–578. https://doi.org/10.1016/j.cub.2014.01.058

D'Aniello, B., Semin, G. R., Alterisio, A., Aria, M., & Scandurra, A. (2018). Interspecies transmission of emotional information via chemosignals: from humans to dogs (Canis lupus familiaris). Animal Cognition, 21(1), 67–78. https://doi.org/10.1007/s10071-017-1139-x

Fiset, S., & Plourde, V. (2015). Commentary: Oxytocin-gaze positive loop and the coevolution of human-dog bonds. Frontiers in Psychology, 6, 1845. https://doi.org/10.3389/fpsyg.2015.01845

Hare, B., Brown, M., Williamson, C., & Tomasello, M. (2002). The domestication of social cognition in dogs. Science, 298(5598), 1634–1636. https://doi.org/10.1126/science.1072702

Müller, C. A., Schmitt, K., Barber, A. L. A., & Huber, L. (2015). Dogs can discriminate emotional expressions of human faces. Current Biology, 25(5), 601–605. https://doi.org/10.1016/j.cub.2014.12.055

Nagasawa, M., Mitsui, S., En, S., Ohtani, N., Ohta, M., Sakuma, Y., Onaka, T., Mogi, K., & Kikusui, T. (2015). Oxytocin-gaze positive loop and the coevolution of human-dog bonds. Science, 348(6232), 333–336. https://doi.org/10.1126/science.1261022

Solomon, J., Beetz, A., Schöberl, I., Gee, N., & Kotrschal, K. (2019). Attachment security in companion dogs: adaptation of Ainsworth's strange situation and classification procedures to dogs and their human caregivers. Attachment & Human Development, 21(4), 389–417. https://doi.org/10.1080/14616734.2018.1517812

Sundman, A.-S., Van Poucke, E., Svensson Holm, A.-C., Faresjö, Å., Theodorsson, E., Jensen, P., & Roth, L. S. V. (2019). Long-term stress levels are synchronized in dogs and their owners. Scientific Reports, 9, 7391. https://doi.org/10.1038/s41598-019-43851-x

Sundman, A.-S., Van Poucke, E., Svensson Holm, A.-C., Faresjö, Å., Theodorsson, E., Jensen, P., & Roth, L. S. V. (2020). Author Correction: Long-term stress levels are synchronized in dogs and their owners. Scientific Reports, 10, 17323. https://doi.org/10.1038/s41598-020-74204-8

Topál, J., Miklósi, Á., Csányi, V., & Dóka, A. (1998). Attachment behavior in dogs (Canis familiaris): a new application of Ainsworth's (1969) strange situation test. Journal of Comparative Psychology, 112(3), 219–229. https://doi.org/10.1037/0735-7036.112.3.219

Wilson, C., Campbell, K., Petzel, Z., & Reeve, C. (2022). Dogs can discriminate between human baseline and psychological stress condition odours. PLOS ONE, 17(9), e0274143. https://doi.org/10.1371/journal.pone.0274143

More on the human-dog relationship

• Building a stronger bond with your dog
• How dogs perceive and respond to our emotions
• Comforting a fearful dog – is it okay?
• Learning theory in dog training
• What is emotion-based dog training?
• Behavior Does Not Equal Emotion
• Attachment in Dogs: what the Strange Situation Test shows and what it doesn’t

Frequently asked questions: dogs as emotion readers

How dogs make our inner state visible – and why that matters so much for training and the relationship